OT: Longhorn strategy a secret?

[Jeremy.Deats]

With the announcement that WinFX will be available for other OS
platforms and that WinFS won't be available at launch, what will
Longhorn offer?

With all this talk about Linux gaining market share, I don't understand
why Microsoft doesn't release Microsoft Linux -or- a release of
Microsoft Windows that runs with Linux under the hood in the same way
MacOS runs with Unix under the hood. What would Microsoft gain from
this? Plenty..

For starters, from a PR persepective much of the security related OS
holes could be washed from Microsoft's hands, they would be in the
hands of the Linux ocommunity. Microsoft could offer exclusive ports of
core products such as Office, IE and server end products such as SQL
Server, IIS and the .NET Framework. All of this could be exclusive to
Microsoft Linux distribution.

Seems like a win win for MS
1. Microsoft gets to make money off Linux customers
2. Microsoft gets to take back market share from other Linux
distributions
3. Microsoft gets positive press from embracing open source
4. That portion of the Linux community that likes to bad mouth
Microsoft Windows would have their soap boxes ripped out from under
them.

If this were to happen (big hypothetical) it would only be a matter of
time before Microsoft became the #1 Linux distribution. If they did
that they would have effectively eaten Linux alive.

 

[Sean Hederman]

With the announcement that WinFX will be available for other OS
platforms and that WinFS won't be available at launch, what will
Longhorn offer?

WinFS will indeed be available, just not in it's "final" form. As for
advantages of one over the other, what advantages did Win2003 offer over
Win2K? Better security, better support for scalable applications, more
stability, better support for .NET applications, and a few other features.
Longhorn will be more of the same, but more so.

With all this talk about Linux gaining market share, I don't understand
why Microsoft doesn't release Microsoft Linux -or- a release of
Microsoft Windows that runs with Linux under the hood in the same way
MacOS runs with Unix under the hood. What would Microsoft gain from
this? Plenty..

For starters, from a PR persepective much of the security related OS
holes could be washed from Microsoft's hands, they would be in the
hands of the Linux ocommunity.

Interestingly Secunia has more security vulnerabilities for Linux than for
Win 2003.
http://blogs.msdn.com/robert_hensing/archive/2005/02/17/375481.aspx

In addition, Windows has a more secure design than Linux. The Linux security
model (like all unixes) is deeply flawed (suid being a prime example) and
does not work well with fine-grained access control, nor with non-file
security. Now if only MS could cut out all the bugs that bypass their
security, they'd be far better off.

Microsoft could offer exclusive ports of
core products such as Office, IE and server end products such as SQL
Server, IIS and the .NET Framework. All of this could be exclusive to
Microsoft Linux distribution.

Do you have the slightest idea of how much work that would be? These
products are absolutely massive, and porting them to Linux would be the work
of years. And for what gain? To have a Microsoft distro contributing to the
already fragmented Linux market? In addition, how exactly would you enforce
that it would only work on the Microsoft distro? The only way to do that
would be to have a distro with so many custom features that it might as well
be a completely separate OS.

Seems like a win win for MS
1. Microsoft gets to make money off Linux customers

Who are not reknowned for spending money on products, and are reknowned for
their hatred of all things Microsoft. Not a very likely selling point.

2. Microsoft gets to take back market share from other Linux
distributions

"Take back" market share? In the desktop world (where most of MS's income
comes from), Microsoft is not losing market share at all. In the server
market, whilst they are losing market share, they're not doing so as fast as
the other Unixes. Most of Linux's gains appear to be coming from
cannabilsation of existing Unix vendors.

3. Microsoft gets positive press from embracing open source

I think most press would have a screaming conniption if Microsoft did this.
The anti-MS press would slate them up and down for "corrupting" the Linux
market. The pro-MS press would have a hissy fit that Microsoft lied to them
about Windows for so many years (since this would be an admission that
Windows was inferior).

4. That portion of the Linux community that likes to bad mouth
Microsoft Windows would have their soap boxes ripped out from under
them.

Nope, as I've indicated before, it would probably make them *more* furious
with MS. Religious hatred does not pay attention to actions, merely to
itself.

If this were to happen (big hypothetical) it would only be a matter of
time before Microsoft became the #1 Linux distribution. If they did
that they would have effectively eaten Linux alive.

The fact of the matter is that Linux still appears to be as far from being
acceptable on the desktop as it was 5 years ago, Windows is slowly starting
to catch up in the security and stability stakes, the Linux market is
fragmenting (http://www.ranum.com/editorials/divide-conquer/) and Microsoft
is in the enviable position of merely having to wait before the whole thing
starts collapsing. They've been here before. Linux would still be my OS of
choice for running a firewall/proxy server, but for LOB applications? Not so
sure. For desktop? Definately not.

In fact my company has actually recently gained a customer who switched to
Linux a few years ago and had such a horriffic experience that they've now
banned all OSS in their organisation (very over the top IMO). An anecdote
does not a trend make, but I'm hearing more and more grumblings in the
industry, from people who work with Linux/Java who are now skilling up on
..NET "just in case". I'm not predicting a sudden downturn in Linuxes
fortunes, but I am predicting a high-water mark in the next few years.

I'd say the biggest threat to MS now is Mac, not Linux (the fact that OSX
runs on Linux is incidental). The Mac mini is a compelling and cheap piece
of hardware, and people have always thought of Mac as "better", especially
when combined with the influence of the iPod. The reason I say that the
Linux base to the Mac OS is incidental is that with the above factors, they
could be using Amiga OS and still be a threat.

 

[Michael S]

Hi Jeremy.
I'm not sure I agree with you at all. Are you perhaps trolling? =)

Anyways, you got my attention and I will give you my 2 cents.

With the announcement that WinFX will be available for other OS
platforms and that WinFS won't be available at launch, what will
Longhorn offer?

I would guess about all the stuff promised without support for WinFX.
Where is the question?

With all this talk about Linux gaining market share, I don't understand
why Microsoft doesn't release Microsoft Linux -or- a release of
Microsoft Windows that runs with Linux under the hood in the same way
MacOS runs with Unix under the hood. What would Microsoft gain from
this? Plenty..

I do understand why Microsoft don't do this. One may can call Windows
bloated or feature-rich, according to taste, but a fact is that Windows is
*HUGE* and it would be an enormous task to port Windows to Linux.

For starters, from a PR persepective much of the security related OS
holes could be washed from Microsoft's hands, they would be in the
hands of the Linux ocommunity. Microsoft could offer exclusive ports of
core products such as Office, IE and server end products such as SQL
Server, IIS and the .NET Framework. All of this could be exclusive to
Microsoft Linux distribution.

I think that Microsoft have already washed their hands regarding security.
IIS 6 and .NET are pretty safe nowadays. With Longhorn security is a number
one priority.

I believe that the few people in the Linux Community that still whines about
Microsoft security holes are, simply put, fundamentalist, that wouldn't be
converted anyways. They don't hate windows, they hate Microsoft as a
corporation.

Seems like a win win for MS
1. Microsoft gets to make money off Linux customers

By having yet another code line that need support and maintainance?

2. Microsoft gets to take back market share from other Linux
distributions

They will? I think Microsoft will do just fine with Longhorn and .NET.

3. Microsoft gets positive press from embracing open source

This is true.

4. That portion of the Linux community that likes to bad mouth
Microsoft Windows would have their soap boxes ripped out from under
them.

In my world, that portion of the linux community is a bunch of whining kids.
While offending them, I am trying to say that they are hardly the ones in a
position of making large IT/IS investments.

If this were to happen (big hypothetical) it would only be a matter of
time before Microsoft became the #1 Linux distribution. If they did
that they would have effectively eaten Linux alive.

Yes, very hypothetical.

Anyways, I don't see a war between Linux and Windows. Developing on .NET
and/or J2EE both have merit.
I believe that these very competent platforms will co-exists. For simple
monolith projects .NET is awesome to work with. For even simpler web
projects LAMP is excellent. For large projects, a fully blown
apache/tomcat/struts/jboss/axis/hibernate/mysql/cvs/maven solution will
ensure quality.

However, after having had to configure the latter I have come to the
conclusion that
- Linx is only free if your time is worthless.

Best regards
- Michael S

 

[Jeremy.Deats]

Regarding the porting of all the products:

"...Do you have the slightest idea of how much work that would be?
These
products are absolutely massive, and porting them to Linux would be the
work
of years. And for what gain? "

Well obviously whatever desktop OS to come after Longhorn is still in
the early stages and I would guess at the 2010-2012 time frame. I'm
assuming by then the majority of Microsoft's product offerings will
have transitioned to managed code, the other compatiblity issues could
be handled by an abstraction layer (e.g. WINE -
http://www.winehq.com/).

 

[Jeremy.Deats]

I don't want to get into a long Windows vs Linux debate, but to be fair
I've seen statisitics that show Linux gaining and losing market share
over the next few years. Many people in influencital positions with
vastly difference opinions. I think our personal experiences can create
a bit of biase. Let's limit the Windows vs Linux debate to this: If
Linux and MacOS continue to gain market share, Microsoft would be
complelled to recover that market share and if they could find a way to
eleminate one of the two through means that already work within their
existing strategies that path may make the most sense.

So we can see Microsoft embracing more open standards. We've had SOAP
for a very long time now, we've had the .NET Framework for a long time
now. We can see XAML and products like Indigo on the horizons, this
technologies are only going to abstract more.

Of course decoupling applications from the OS is a enormous task, but
I think we've seen Microsoft embracing this strategy for many years
now.

Most consumers don't understand technology at a detail level, but the
personal computer has been with us for almost a generation now. It's
kind of startling that only one primary OS dominates so much market
share. I think customers are feeling comfortable enough with the
technology to start looking at alternatives and around Windows we've
had two extreems, MacOS and Linux, but MacOS is becoming less expensive
and more powerful (in terms of allowing users to do what they want to
do with a desktop PC) and Linux is becoming more user friendly.
There's no good analogy, but this one comes somewhat close: So the way
it's been consumers have a choice between a Honda Accord (Windows) a
BMW 3-Series (MacOS) or Ford Mustang GT stock car package (Linux). The
line between the three has been blured.and continues to become more
blured. As that continues to happen the choice of the consumer becomes
more difficult and that will translate into a loss of Microsoft's
market share and a gain for the other two (and I agree MacOS probably
stands the most to gain).

The other side of this coin is the rise of open source products and
consumer acceptance of open source products. I think FireFox has proven
Windows customers are become more accepting of open source
applications. If it's a solid application that brings new features to
the table, the average customer probably doesn't care where it comes
from. So there are projects such as Mono that could gain ground and
also help push Microsoft in the direction of Linux.

 

[Sean Hederman]

Regarding the porting of all the products:

"...Do you have the slightest idea of how much work that would be?
These
products are absolutely massive, and porting them to Linux would be the
work
of years. And for what gain? "

Well obviously whatever desktop OS to come after Longhorn is still in
the early stages and I would guess at the 2010-2012 time frame. I'm
assuming by then the majority of Microsoft's product offerings will
have transitioned to managed code, the other compatiblity issues could
be handled by an abstraction layer (e.g. WINE -
http://www.winehq.com/).

Have you ever tried actually using WINE? It's slow and doesn't implement
quite a bit of Win32. The fact of the matter is that MS is going to keep
adding functionality to the underlying OS and abstraction layers and
framework ports are going to have to run to keep up. In any case, my
original point still holds, there's no advantage to porting them to Linux,
..NET maybe, but not Linux. Once all the MS products run on Linux and there
is a fully fledged .NET framework on Linux which supports all the MS
additions, then the porting to Linux will occur automatically. I don't
expect Microsoft to be terribly interested in helping that come about
though.

 

[Sean Hederman]

I don't want to get into a long Windows vs Linux debate, but to be fair
I've seen statisitics that show Linux gaining and losing market share
over the next few years. Many people in influencital positions with
vastly difference opinions. I think our personal experiences can create
a bit of biase. Let's limit the Windows vs Linux debate to this: If
Linux and MacOS continue to gain market share, Microsoft would be
complelled to recover that market share and if they could find a way to
eleminate one of the two through means that already work within their
existing strategies that path may make the most sense.

For starters we're assuming that these OS's will continue to gain market
share indefinately, which as I've indicated, I have some doubts about. In a
scenario where a "perfect storm" of sudden Linux desktop acceptance and
massive Mac gains occurs, then I do agree that MS would have to start
worrying. Such a scenario, even in the most optimistic scenarios, is still
years away, so there's not much point in MS planning for it now. Especially
if there's a pretty good chance that such a thing will not come to pass.

So we can see Microsoft embracing more open standards. We've had SOAP
for a very long time now, we've had the .NET Framework for a long time
now. We can see XAML and products like Indigo on the horizons, this
technologies are only going to abstract more.

I agree that embracing open standards is often very helpful (especially to
us poor schmucks doing cross-platform integration), and I think MS is doing
good work in this. I think that their main reason for this is that they've
realised that other OS's will be around for years to come, and if they don't
play nice then people will switch. As for abstracting technologies further
and further, at some point the abstractions become so divorced from reality
that they lose all usability
(http://www.joelonsoftware.com/articles/fog0000000018.html).

Of course decoupling applications from the OS is a enormous task, but
I think we've seen Microsoft embracing this strategy for many years
now.

Decoupling applications from the OS is virtually impossible. What .NET is
allowing us to do is the same as Java: provide a "logical" OS that can in
theory be implemented anywhere. Said applications are still coupled to that
virtual platform however, and if that virtual platform has services only
available on Microsoft Windows...

Most consumers don't understand technology at a detail level, but the
personal computer has been with us for almost a generation now. It's
kind of startling that only one primary OS dominates so much market
share.

Not really, it actually makes a lot of sense. People want systems that
easily interoperate with their friends, families and work systems. Similarly
businesses want systems that interoperate with their customers, vendors and
employees systems. Given that, we can easily see that once an OS started
becoming big in any of these spheres it would quickly "spread" to the
others. I think the surprise was that nobody really thought it would be the
home PC wagging the dog.

I think customers are feeling comfortable enough with the
technology to start looking at alternatives and around Windows we've
had two extreems, MacOS and Linux, but MacOS is becoming less expensive
and more powerful (in terms of allowing users to do what they want to
do with a desktop PC) and Linux is becoming more user friendly.

And both have made great strides in interoperating with Windows, which is
perhaps more important. Such interoperability allows for lower barriers to
switching operating systems. It remains to be seen if Apple can trade on
their good name enough to make significant inroads into the desktop market.
Similarly, I've been hearing about how Linux is going to take off in that
space for years now with no appreciable follow-through. Maybe this year,
maybe next, maybe in five years, maybe never.

There's no good analogy, but this one comes somewhat close: So the way
it's been consumers have a choice between a Honda Accord (Windows) a
BMW 3-Series (MacOS) or Ford Mustang GT stock car package (Linux). The
line between the three has been blured.and continues to become more
blured. As that continues to happen the choice of the consumer becomes
more difficult and that will translate into a loss of Microsoft's
market share and a gain for the other two (and I agree MacOS probably
stands the most to gain).

The blurring you're talking about will only really take effect once there is
one single platform to develop against. Whether this will be Java or .NET or
something else remains to be seen.

The other side of this coin is the rise of open source products and
consumer acceptance of open source products. I think FireFox has proven
Windows customers are become more accepting of open source
applications. If it's a solid application that brings new features to
the table, the average customer probably doesn't care where it comes
from. So there are projects such as Mono that could gain ground and
also help push Microsoft in the direction of Linux.

None of these open source projects has yet gained serious market share on
the desktop space. Firefox looks the most promising, but a lot of users
switched to it as they believed it would be inherently more secure. If more
major Firefox vulnerabilities appear acceptance might slow down drastically.
Keep in mind that another reason Firefox was gaining was due to the lack of
interest MS showed in updating IE. If they put a serious effort into this,
they might be able to take back some of the Firefox market share. As for
Mono, I doubt it will "push" Microsoft anywhere. MS will continue to extend
and update .NET, and gradually port products to .NET. If some of those
products happen to work on Mono, great. If not, they won't care. In fact I
think the biggest advantage of Mono to MS will be to allow them to more
easily and cheaply provide Office for Mac.

In any case, I personally don't think that moving to Linux is the panacea
OSS advocates claim it will be.
I've blogged a bit on my feelings about OSS versus CSS here:
http://codingsanity.blogspot.com/2005/03/is-oss-better.html

 

[Michael S]

Very intresting thread. I apologize for calling you a troll Jeremy.

Regards
- Michael S